“This document is the Accepted Manuscript version of a Published Work that appeared in final form in ACS Catal. 2017, 7,
2367-2373, copyright © American Chemical Society after peer review and technical editing by the publisher. To access the
final edited and published work see DOI: 10.1021/acscatal.7b00109. This article may be used for non-commercial purposes
in accordance with the ACS guidelines published at http://pubs.acs.org/page/policy/articlesonrequest/index.html].”

1

Vanadium(V) Catalysts with High Activity for the
Coupling of Epoxides and CO2: Characterization of
a Putative Catalytic Intermediate
Claudia Miceli,§ Jeroen Rintjema,† Eddy Martin,† Eduardo C. Escudero-Adán,† Cristiano
Zonta,§ Giulia Licini,*,§ and Arjan W. Kleij*,†,‡

§

Dipartimento di Scienze Chimiche and CIRCC, Università di Padova, Via Marzolo 1, 35131 –
Padova, Italy
†

Institute of Chemical Research of Catalonia (ICIQ), the Barcelona Institute of Science and
Technology (BIST), Av. Països Catalans 16, 43007 – Tarragona, Spain.

‡

Catalan Institute of Research and Advanced Studies (ICREA), Pg. Lluís Companys 23, 08010 –
Barcelona, Spain.

2

ABSTRACT: Vanadium(V) complexes derived from aminotriphenolate ligands are
demonstrated to be highly active catalysts for the coupling of various terminal and internal
epoxides with carbon dioxide to afford a series of substituted organic carbonates in good yields.
Intriguingly, a V(V) complex bearing peripheral chloride groups on the ligand framework allowed
for the formation and isolation of a rare complex that incorporates a ring-opened epoxide with one
of the phenolateO atoms acting as a nucleophile and the metal center as a Lewis acidic site. This
unusual structure was characterized by X-ray diffraction and 51V-NMR, and was shown to exhibit
catalytic activity for the coupling of propylene oxide and CO2 when combined post-synthetically
with these substrates. The results obtained herein clearly show that vanadium complexes in a high
oxidation state are powerful catalysts for the activation of challenging internal epoxides, and their
conversion into cyclic organic carbonates.

Keywords: aminotriphenolate ligands, carbon dioxide, cyclic carbonates, homogenous
catalysis, vanadium

3

 INTRODUCTION
Small molecule activation, and particularly carbon dioxide (CO2) capture and conversion,1
continues to challenge the synthetic communities to devise more efficient and sustainable catalysis
protocols for classically low-reactive molecules. In the realm of sustainable development, it is
essential to design new catalyst systems that are characterized by an easy, modular and costeffective construction while addressing the requisite of general availability of the raw materials
and using thus abundant and preferably renewable chemical building blocks.2 Carbon dioxide
represents an interesting source of carbon in synthetic chemistry, but its successful conversion
relies on the use of catalysts and high-energy co-reactants that can mediate its conversion, since
the latter is challenged both from a kinetic and thermodynamic point of view.3 Over the last decade,
the growth of new catalytic technologies for CO2 conversion has been spectacular with progress
reported for a wide series of transformations.4 Metal catalysis has been prominent in most of the
synthetic efforts directed towards CO2 “fixation” activating either this carbon feedstock directly
or a coreactant able to induce a coupling reaction.5 An often applied strategy in this regard has
been the Lewis acid activation of small heterocyclic ethers in the presence of a nucleophilic
additive giving rise to alkoxide intermediates that smoothly react with CO2 to form carbonate
structures: these linear carbonate species eventually evolve into either cyclic6 or polycarbonates7
and the observed chemoselectivity typically depends on the reaction conditions, the type of
nucleophile and Lewis acidic metal complex.
Successful catalysts for cyclic organic carbonate (COC) formation include the use of abundant
main group metals such as aluminum,8 iron,9 and zinc,10 and transition metals (Co, Cr)11 most often
in combination with Nx,Oy-coordinating ligands such as salens (x = y = 2)12 and aminotriphenolates

4

(x = 1, y =3).13 We have been interested in the use of O3N-chelating aminotriphenolate ligands and
their complexes for catalytic applications, and have reported on the use of Ti(IV) derivatives for
the oxidation of sulfides and secondary amines into nitrones,14 Mo(VI) and W(VI) complexes for
the oxidation of sulfides, olefins and chloro and bromoperoxidation,15 and more recently the
Fe(III) mediated formation of cyclic carbonates and the preparation of Catechol Oxidase and
Catalase mimics.16 The use of aminophenolate type ligands is beneficial towards the stabilization
of different (higher) oxidation states such as V(III) and V(V),17 and the V(V) complexes have
previously been successfully used in the context of haloperoxidase models in sulfoxidation
chemistry.18 Inspired by the presence of high-oxidation state V centers in these complexes
representing strongly Lewis acidic sites, we envisioned that such compounds would also be highly
efficient in epoxide activation/ring-opening in the presence of a suitable nucleophile, and give rise
to COC formation in the presence of CO2. As far as we know, the use of vanadium in the formation
of COCs and related compounds remains rare, and with limited application potential.19

Figure 1. V(V) aminotriphenolate complexes 1 and 2 used in this work as catalyst components for
COC formation.

Therefore, we set out to explore the potential of V(V) aminotriphenolate complexes (Figure 1)
in the formation of COC and have specifically focused on the conversion of more challenging sub-

5

series of internal epoxide partners. The use of this class of substrates is currently attracting more
attention as efficient catalysts have become available for their conversion,20 and therefore new
catalyst systems should be comparatively evaluated against this advanced state of the art in COC
formation. In this work we will demonstrate that V(V) aminotriphenolate complexes combined
with suitable halide nucleophiles offer highly efficient binary catalysts for COC formation from a
wide series of internal epoxides. Further to this, the unexpected low activity for the conversion of
some terminal epoxides by the strongest Lewis acidic V(V) complex was investigated in more
detail. This has resulted in the formation and structural characterization of a unique and
unprecedented catalytic intermediate where the V(V) center and one of the phenolate donors had
cooperatively activated the epoxide substrate to form a six-coordinate V(V) complex comprising
of a bidentate aryloxo-alkylether fragment. This new epoxide activation mode was confirmed by
X-ray analysis, and further use of this intermediate in COC formation was also demonstrated. Our
results thus demonstrate the use of a powerful new binary catalyst system for challenging coupling
reactions between internal epoxides and CO2, and mark the use of abundant vanadium as a useful
and sustainable alternative in COC formation.

 RESULTS AND DISCUSSION
2.1 Screening Phase. The V(V) complexes 1 and 2 were combined with NBu4I and used as
binary catalysts for the formation of COCs DF from terminal epoxides AC (Table 1). As a
reference catalyst, we used the previously reported and highly active binary catalyst 3/NBu4I21
under similar conditions to examine the catalytic potential of the V(V) complexes.

6

Table 1. Screening Experiments carried out with V(V) Complexes 1 and 2 using various
Terminal Epoxides (AC) and one Internal Epoxide (D).a

Entry

substrate

Cat. (%)

Nu (%)

T (ºC)

Yield (%)b

1
1, 0.50
1.25
45
>99
A
2
1, 0.50
45
0
A

3
2, 0.50
45
0
A

4
1.25
45
64
A

5
1, 0.20
0.50
45
94
A
6
1, 0.10
0.25
45
58
A
7
1, 0.10
45
0
A

8
0.25
45
23
A

9
2, 0.10
0.25
45
73
A
10
3, 0.10
0.25
45
>99
A
11
1, 0.10
0.25
45
70
B
12
2, 0.10
0.25
45
19
B
13
3, 0.10
0.25
45
74
B
14
0.25
45
7
B

15
1, 0.10
0.25
45
67
C
16
2, 0.10
0.25
45
18
C
17
3, 0.10
0.25
45
60
C
18
0.25
45
26
C

c
19
2, 0.50
1.25
85
70d
D
20c
2, 0.50
5.0
85
>99 (68)e
D
c
21
1, 0.50
5.0
85
93 (41)e
D
c
22
2, 0.25
5.0
85
>99
D
23c
5.0
85
26
D

c
24
3, 0.50
5.0
85
>99 (55)e
D
a
Reaction conditions: 2.0 mmol of epoxide, temperature indicated, NBu4I (amount indicated), 10
bar, 18 h, neat. Nu stands for nucleophilic additive, and catalyst and nucleophile loadings are in
mol%. bNMR yield using mesitylene as internal standard; selectivity towards the COC product
>99%. cConditions used: 1.0 mmol D, NBu4Br (amount indicated), 10 bar, 18 h, neat. dConversion
was 98%, with benzaldehyde (28%) being formed as a by-product. eIn parentheses the NMR yield
after 6 h.

7

The conversion of 1,2-epoxyhexane A was first evaluated and a combination of complex 1 (0.50
mol%) with NBu4I (1.25 mol%) provided quantitative yield for the COC product D (entry 1). The
complexes 1 and 2 themselves proved to be unproductive (entries 2 and 3) whereas NBu4I alone
gave a 64% yield of D. This prompted us to further lower the loading of 1 and NBu4I (entries 57)
and we were pleased to find that a combination of 1 (0.10 mol%) and NBu4I (0.250.50 mol%)
still provided appreciable to high yield of the COC E whereas the yield in the presence of only
NBu4I was significantly lower (entry 8). Upon using the more Lewis acidic V(V) complex 2 (cf.,
entries 6 and 9, 58% vs. 73%) a higher yield of E was noted compared to the binary system
1/NBu4I. Comparison with the previously reported and highly active binary catalysts 3/NBu4I
(entry 10),5f,21 shows that for 1,2-epoxyhexane conversion the binary systems based on Vcomplexes 1 and 2 have lower, though still appreciable activity. Other terminal epoxides were then
also examined (styrene oxide B and epichlorohydrin C; entries 1118), and surprisingly in these
cases the binary catalyst based on 1 proved to be significantly more active than the one derived
from 2. Catalyst 1/NBu4I showed again a considerably higher yield for COCs F and G compared
to the background reaction mediated by the nucleophile only but, more importantly, similar order
magnitude COC yields as determined for the Al-based binary system 3/NBu4I (cf., entries 11 vs.
13 and 15 vs. 17). Since binary 3/NBu4I catalysts have been reported among the most active for
COC formation using terminal epoxides,5f,21 it therefore seems that aminotriphenolate V(V)
complexes have also high catalytic potential.
In order to challenge the binary couples consisting of 1 or 2 and a suitable ammonium halide
nucleophile towards COC formation, an internal epoxide (trans D) was then probed as a reaction
partner (entries 1924). Interestingly, in the formation of COC H the use of complex 2 was more
beneficial showing higher yield of the desired product (cf., entries 20 and 21). At a loading of 1.25

8

mol% of NBu4Br,22 however, substantial formation of benzaldehyde (entry 19, 28%) was noted as
a likely result of the V(V) complex acting as a catalyst for oxidative cleavage after initial formation
of a diol intermediate.18c An increase in the nucleophile loading up to 5.0 mol% solved this
problem with the chemo-selectivity, and under these conditions (entry 20) quantitative formation
of COC H was obtained, and for this substrate upon using Al-complex 3 (entry 24) a slightly lower
yield was obtained after a 6 h reaction time. When comparing the V-based catalysts of this work
with a recently reported diphenolate V(V) based complex (0.20 mol%) combined with NBu4Br
(0.20 mol%) for the conversion of PO at 80ºC and 40 bar CO2 pressure (18 h; conversion 79%),19c
it seems that the complexes of this study show relative good potential. However, since some
disparity exists between our conditions (Table 1) and the ones reported previously, some caution
is needed to quantitatively compare these data.23 The conditions reported for the successful
formation of disubstituted COC H using V-complex 2 were then taken as a starting point for the
coupling of various internal epoxides with CO2.

2.2 Scope in Disubstituted COC products. Motivated by the results summarized in Table
1, a series of internal epoxides was then converted into their disubstituted COCs 419 (see Figure
2) using the binary catalyst 2/NBu4Br. The product diversity proved to be wide, and various
(functional) epoxides (either commercial ones or prepared in few steps from easily accessible
starting materials, see the Supporting Information; SI) could be converted in their cyclic carbonates
typically in good yields. Different ring-size epoxides are suitable substrate partners including five, six- and eight-membered ones (49, 17-19), though in the preparation of 7 and 9 we observed
that longer reaction times (up to 3 days) and/or increasing the binary catalyst loading did not
improve the results. COC 7 is a rather interesting structure that comprises an epoxide unit that was
unaffected during the catalytic trials, and thus remains available for post-synthetic modification.

9

The binary catalyst 2/NBu4Br also proved to be useful to convert a trimethoxysilylalkyl derived
epoxide without observable degradation (cf., formation of 10). COC product 11 was obtained from
a biomass-derived fatty acid precursor (oleic acid) and represents an interesting contextual case
towards biomass valorization.24 The COCs 1316 were prepared from their stereo-pure cis or trans
precursors: whereas the coupling of trans 2,3-dimethyloxirane with CO2 proceeded with high
retention of configuration (13: cis/trans = 4:96), the use of the diastereo-isomerically pure cis
epoxide led to substantial loss of stereo-information

10

Figure 2. Synthesis of disubstituted COCs 419 from internal epoxides and CO2 using the binary
catalyst 2/NBu4Br. General conditions: epoxide (4 mmol), 2 (0.5 mol%), NBu4Br (5 mol%), neat,
85 °C, p(CO2)º = 10 bar, 18 h. The dr values were determined by 13C NMR spectroscopy using
inverse-gated decoupling, and reported yields are isolated ones after column purification. Reported
relative configurations of the drawn structures refer to the major/expected product. [a] In these
reactions an aldehyde by-product was formed as shown by 1H NMR.

(14: cis/trans = 52:48). Such loss of stereo-information was reported before and is generally
explained by the occurrence of both SN1 and SN2 based ring closing mechanisms.20e,24d The use of
a sterically more congested substrate (cis/trans 2,3-diphenyloxirane), however, led to exclusive
isolation of the trans-configured products (15 and 16) regardless the initial configuration of the
epoxide substrate. The combined results clearly show the high catalytic potential of V(V) complex
2 in the formation of these challenging conversions, and the scope of COC products reported in
Figure 2 is among the widest reported to date.

2.3 Mechanistic Investigations. Based on the remarkable and relatively low activity found
for the activation of styrene oxide and epichlorohydrin (Table 1) by the more Lewis acidic V(V)
complex 2 compared to 1, we decided to investigate the origin of this behavior. Since previous
work on related Al(III) aminotriphenolate complexes showed that cyclic ether coordination5f,21
precedes the ring-opening process by an external nucleophile, we first combined complex 2 and
the simplest terminal epoxide (propylene oxide, PO) and focused on the formation of a similar
epoxide-coordinated structure and its analysis by X‒ray diffraction. Fortunately, crystalline
material obtained from a mixture of 2 and excess PO in CH3CN could be easily obtained and the
determined structure (2PO) is presented in Figure 3. Much to our surprise, instead of the expected

11

PO-coordinated V(V) complex,25 unprecedented formation of a V(V) complex comprising of a
ring-opened PO had occurred mediated by one of the phenolate Oatoms that acted as an internal
nucleophile.
The V-center in 2‒PO is asymmetrically positioned in a rather distorted octahedral coordination
structure, with the vanadyl fragment (V(1)O(1) = 1.5877 Å) occupying one of the equatorial
positions. The three Odonors of the aminophenolate ligand occupy the remaining positions of the
equatorial plane, with the V(1)O(3) distance (2.3413 Å) being significantly longer than the other
VO bonds (V(1)O(2) and V(1)O(5) of 1.8912 and 1.8701 Å, respectively). The axial Odonor
originating from the activated epoxide shows a relatively short bond length of 1.7871 Å
(V(1)O(4)) pointing at negative charge localization at this Ocenter, and thus the organic
fragment connected through O(3) and O(4) should be regarded as a mono-anionic arylethanolate
fragment with V(1)O(3) being a dative bond involving an arylalkyl ether donor.

12

Figure 3. X-ray molecular structure of the unusual POadduct 2PO based on V(V) complex 2.
Selected bond lengths (Å) and angles (º), with esd´s in parentheses: V(1)-O(1) = 1.5877(17), V(1)O(2) = 1.8912(16), V(1)-O(3) = 2.3413(17), V(1)-O(4) = 1.7871(17), V(1)-O(5) = 1.8701(16),
V(1)-N(1) = 2.244(2); O(4)-V(1)-N(1) = 155.88(8), O(1)-V(1)-O(3) = 178.71(8), O(2)-V(1)-O(5)
= 158.95(7).

13

Thus, it seems reasonable to suggest that after initial coordination of the PO molecule, both the
Lewis acidic V center and one of the Lewis basic phenolate cooperatively activate the epoxide and
form a stable, six-coordinate V(V) complex. Interestingly, such a process shows parallels with the
ring-opening of the epoxide by an external nucleophile (often proposed to be the first step in COC
formation using binary catalysts), though in the present case the phenolate acts as an internal
nucleophile. The structure of 2PO was further analyzed/confirmed by MALDI-TOF mass
analysis and 51V-NMR spectroscopy.
To gain further insight in the formation of 2PO,

51

V NMR titration studies were carried out

using 2 and different amounts of PO (Figure 4). The respective NMR traces were recorded 5 h
after initial mixing (traces bf), and a comparison was made with the 51V resonances for 2 (trace
a) and isolated 2PO (trace g). Upon addition of PO, a multifold of resonances appeared that seem
to converge to two major resonances located at 457 and 470 ppm. The latter peak assignment
was facilitated by the 51V NMR analysis of an isolated crystalline sample of 2PO and could thus
be assigned to this unusual structure. Furthermore, the reaction of 2 and PO to induce the formation
of 2PO depends on the excess of PO present in solution; interestingly, at a lower excess of PO
(33 equiv, d) the spectrum evolves in time (i.e., a few days) to the same final trace observed under
(f) suggesting that full conversion of 2 into 2PO under these conditions (25ºC, 33fold excess of
PO) is relatively slow.
It is likely that the formation of 2PO is preceded by simple PO coordination which may occur
both in an equatorial and axial fashion. Also the final in situ prepared product 2PO could have
the vanadyl fragment occupying either an axial or equatorial position in the octahedral complex,
which could explain the presence of the two main peaks. Such behavior was also demonstrated for
other V(V) complexes comprising of O2,N,N´-coordinated tetradentate ligands showing V-

14

resonances in the same region as noted for 2PO.17b,18b Alternatively, in solution Lewis acidic
2PO may be part of larger clusters such as dimers that can affect the 51V resonance and account
for the presence of some other minor species (cf., trace f).

-431.1

(a)

+ 10 equiv

(b)

+ 20 equiv

(c)

+ 33 equiv

(d)

+ 66 equiv

(e)

+ 100 equiv

-456.9

(f)
-470.4

Figure 4.

51

(g)

V-NMR titration of complex 2 with various amounts of PO. Conditions used: [2] =

0.01 M in CD3CN, NMR spectrum was recorded 5 h after initial mixing of 2 and PO. See the SI
for further details.

In order to test whether the pre-isolated complex 2PO acts as a functional model for the epoxide
ring-opening stage, we performed some further catalytic tests with a pre-isolated sample of 2PO

15

and combined it with NBu4Br26 and fresh PO, and used it as a binary system for the formation of
propylene carbonate (see Table 2). The catalysis data was compared against the background
reaction (use of NBu4Br only, entry 1), and the activity of 2/NBu4Br and 1/NBu4Br (entries 2 and
3) under similar reaction conditions and catalyst loadings. From the data in Table 2 it is clear that
the complex 2PO still exhibits catalytic activity toward COC formation as it delivers propylene
carbonate in a significantly higher yield than the background reaction. Remarkably, the
combination of parent V(V) complex 2 with NBu4Br under the reported conditions and (low)
catalyst loading seems to possess similar activity as the binary combination 2PO/NBu4Br
pointing towards a relatively fast in situ formation of 2PO at 50ºC and >800 equiv of PO (Table
2).

Table 2. Comparison of the Activity of the Binary Systems 2-PO/NBu4Br, 1/NBu4Br and
2/NBu4Br in the Formation of Propylene Carbonate.a

Entry

V-complex (mol %)

Nu (mol%)

Yield (%)b

1
NBu4Br, 0.048
3

2
1, 0.024
NBu4Br, 0.048
96
3
2, 0.024
NBu4Br, 0.048
9
4
2-PO, 0.024
NBu4Br, 0.048
9
a
Reaction conditions: 20.0 mmol propylene oxide, 50ºC, 10 bar, 18 h, neat. Nu stands for
nucleophilic additive. bNMR yield determined using mesitylene as internal standard; selectivity
towards the COC product >99%. The average of two runs is reported.

16

As noted for the conversion of styrene oxide B and epichlorohydrin C (Table 1), a much lower
activity for COC formation is noted when compared with the results obtained with the binary
system 1/NBu4Br, suggesting that terminal epoxides with a modest steric impediment likely have
a higher tendency to (partially) form an epoxide adduct of type 2PO. Further evidence for this
hypothesis was gathered by analysis of the resultant complex when combining 2 with 100 equiv
of epichlorohydrin (ECH) and analyzing the product by

51

V NMR and MALDI-TOF mass

spectrometry (see SI), and Xray diffraction (Figure 5).

Figure 5. X-ray molecular structure of 2ECH based on V(V) complex 2, ECH = epichlorohydrin.
Selected bond lengths (Å) and angles (º), with esd´s in parentheses: V(1)-O(1) = 1.585(4), V(1)O(2) = 1.893(4), V(1)-O(3) = 2.327(4), V(1)-O(4) = 1.879(4), V(1)-O(5) = 1.808(4), V(1)-N(1) =
2.233(5); O(4)-V(1)-N(1) = 84.25(17), O(1)-V(1)-O(3) = 179.8(2), O(2)-V(1)-O(5) = 92.21(19),
O(2)-V(1)-O(4) = 159.53(18), N(1)-V(1)-O(5) = 156.53(19).

17

Similar formation of an unusual epoxide adduct was noted (cf., formation of 2‒ECH), and therefore
the combined analytical data clearly show a more general tendency to form intermediates of type
2PO when using terminal epoxide substrates with relatively small substituents.
In order to further test this hypothesis, crystallization of the sterically more crowded V(V)
complex 1 was carried out in the presence of PO: Xray diffraction studies revealed that the
crystals corresponded to the starting complex 1 with no additional ligand being present in the
coordination sphere (X‒ray structure determined, see SI).27 This seems to suggest that the
coordination of PO and its subsequent incorporation into the V(V) aminotriphenolate complex as
noted in 2PO is sterically modulated, but electronic effects cannot be ruled out as the V center
may be involved in activating the epoxide towards intramolecular nucleophilic attack by one of
the phenolate O-atoms. Furthermore, upon using sterically more hindered/demanding epoxides
such as 1,2-epoxyhexane A and trans 2-phenyl-3-methyloxirane D, the expected reactivity order
is restored. The more Lewis acidic complex 2 shows in these cases higher activity upon combining
with NBu4X (X = I, Br; Table 1) than noted for 1 under comparable conditions. These latter
observations thus justify the use of the more Lewis acidic complex 2 in the conversion of internal,
sterically more congested epoxides and the formation of COCs 4‒19 (Figure 2).
Based on our observations and analyses, we propose that the formation of COCs using 2 starts
off with coordination of the epoxide giving rise to a hexa-coordinated V(V) complex (Figure 6).
Such coordination chemistry has been previously observed for various V(V) complexes ligated by
tetradentate ligands in the literature17a-b,18b and recently also a H2O-ligated V(V) aminotriphenolate
was crystallographically characterized.17a In the presence of terminal epoxides, the formation of a
complex of type 2‒PO can occur with a higher tendency if the epoxide substituent is relatively
small.28 It should be noted that the epoxide-based structures themselves (2‒PO and 2‒ECH) do not

18

show any catalytic turnover and require the addition of an external nucleophile for reactivation.
Upon reactivation, complex 2 can reenter the catalytic cycle in the presence of NBu4X (X = Br, I),
suggesting the subsequent formation of V(V) alkoxide intermediate species with the halide
nucleophile X incorporated. Such a species can be formed through a nucleophilic attack of the
bromide onto the arylalkyl ether part of the unusual intermediate, and allows for a nucleophilic
displacement of the original phenolate. It is likely that this step is turnover-limiting in those cases
where such a stable intermediate (cf., 2‒PO and 2‒

Figure 6. Proposed mechanistic manifold for the formation of COCs using the binary catalyst
2/NBu4X (X = Br, I).

19

ECH) can be formed, and therefore explains the lower relative reactivity for 2/NBu4I in the
conversion of terminal epoxides with smaller substituents. Once the halide-containing alkoxide
intermediate is formed, the formation of the COC follows a standard CO2 insertion and ringclosing sequence delivering the final product and regenerating the five-coordinated V(V) complex
2 and NBu4X for further catalytic turnover.

 CONCLUSION
In this contribution we show that vanadium (V) complexes derived from aminotriphenolate
ligands form powerful binary catalysts with ammonium salts and allow for the conversion of
generally more challenging internal epoxides and CO2 into their respective COCs with ample
scope in epoxide reaction partners that display functional group diversity. Furthermore, the
screening data and mechanistic control reactions together with the Xray molecular structures
determined for 2PO and 2ECH have revealed a new manifold towards the activation of the
epoxide substrate leading to the formation of a functional model of an alkoxide-related
(epoxideactivated) intermediate. The complexes 2PO and 2ECH can reengage in COC
formation after combining with an ammonium salt. The unprecedented formation of these unusual
compounds is sterically regulated, and has been confirmed to be operative for epoxides with
limited steric impediment (propylene oxide, epichlorohydrin), whereas for sterically more
encumbered substrates it can be assumed that initial coordination of the epoxide to the V(V) center
is followed by epoxide ring-opening mediated by an external (halide) nucleophile. Therefore this
work does not only show the high potential of high oxidation state V-complexes as binary catalyst
components in COC formation, but the use of a V(V) aminophenolate species has also allowed for

20

the first time to identify a unique epoxide-activation mode reminiscent of a generally proposed
intermediate in COC formation following epoxide activation by a Lewis acid and an external
nucleophile.

AUTHOR INFORMATION
Corresponding Authors
*E-mail for G.L: giulia.licini@unipd.it.
*E-mail for A.W.K.: akleij@iciq.es.
Orchid
Arjan W. Kleij: 0000-0002-7402-4764
Notes
The authors declare no competing financial interest.

 ASSOCIATED CONTENT
The Supporting Information is available free of charge on the ACS Publications website at DOI:
10.1021/acscatal. ####.
Characterization data for all the organic carbonates and complexes, and
crystallographic details in cif format (PDF).

 ACKNOWLEDGMENTS
We thank ICREA, the CERCA Program/Generalitat de Catalunya and the Spanish Ministerio de
Economía y Competitividad (MINECO: CTQ-2014-60419-R and Severo Ochoa Excellence

21

Accreditation 2014−2018, SEV-2013-0319), the Università di Padova (PRATCPDA123307) and
MIUR (Progetto Competitivo CMPT148505) for financial support. J.R. thanks the ICIQ
foundation for a predoctoral fellowship, and C. M. acknowledges the COST action CM1205
(CARISMA) for an STSM fellowship. We also thank Dr. Noemí Cabello for the MS studies.

REFERENCES
(1) For some general reviews: (a) Aresta, M.; Dibenedetto, A.; Angelini, A. Chem. Rev. 2014,
114, 17091742. (b) Peters, M.; Köhler, B.; Kuckshinrichs, W.; Leitner, W.; Markewitz, P.;
Müller, T. E. ChemSusChem 2011, 4, 12161240. (c) Sanz-Pérez, E. S.; Murdock, C. R.; Didas,
S. A.; Jones, C. W. Chem. Rev. 2016, 116, 1184011876. (d) Kondratenko, E. V.; Mul, G.;
Baltrusaitis, J.; Larrazábal, G. O.; Pérez-Ramírez, J. Energy Environ. Sci. 2013, 6, 31123135.
(2) (a) Renewable Raw Materials – New Feedstocks for the Chemical Industry; Ulber, R.; Sell,
D.; Hirth, T.; Eds.; Wiley-VCH: Weinheim, 2011, p 1-244. (b) Ciriminna, R.; Lomeli-Rodriguez,
M.; Demma Carà, P.; Lopez-Sanchez, J. A.; Pagliaro, M. Chem. Commun. 2014, 50, 15288-15296.
(c) Williams, C. K.; Hillmyer, M. A. Polym. Rev. 2008, 48, 1-10. (d) Winkler, M.; Romain, C.;
Meier, M. A. R.; Williams, C. K. Green Chem. 2015, 17, 300-306.
(3) (a) Carbon Dioxide as Chemical Feedstock; Aresta, M.; Ed.; Wiley-VCH: Weinheim, 2010,
p. 1-414. (b) Sakakura, T.; Choi, J. C.; Yasuda, H. Chem. Rev. 2007, 107, 2365-2387. (c)
Perathoner, S.; Centi, G. ChemSusChem 2014, 7, 1274-1282.
(4) General reviews on CO2 transformations: (a) Klankermayer, J.; Wesselbaum, S.; Beydoun,
K.; Leitner, W. Angew. Chem. Int. Ed. 2016, 55, 7296-7343. (b) Liu, Q.; Wu, L.; Jackstell, R.;

22

Beller, M. Nat. Commun. 2015, 6, 5933. (c) Cokoja, M.; Bruckmeier, C.; Rieger, B.; Herrmann,
W. A.; Kühn, F. E. Angew. Chem. Int. Ed. 2011, 50, 8510-8537. (d) Yu, B.; He, L.-N.
ChemSusChem 2015, 8, 52-62. (e) Kielland, N.; Whiteoak, C. J.; Kleij, A. W. Adv. Synth. Catal.
2013, 355, 2115-2138.
(5) For selected, seminal examples of direct CO2 activation: (a) Aresta, M.; Nobile, C. F.;
Albano, V. G.; Forni, E.; Manassero, M J. Chem. Soc. Chem. Commun. 1975, 636-637. (b)
Fachinetti, G.; Floriani, C.; Zanazzi, P. F. J. Am. Chem. Soc. 1978, 100, 7405-7407. (c) CastroRodriguez, I.; Nakai, H.; Zakharov, L. N.; Rheingold, A. L.; Meyer, K. Science 2004, 305, 17571759. For a small selection of generally applicable approaches of co-reactant (H2, epoxide,
propargylic systems) activation, see: (d) Tanaka, R.; Yamashita, M.; Nozaki, K. J. Am. Chem. Soc.
2009, 131, 14168-14169. (e) Wesselbaum, S.; vom Stein, T.; Klankermayer, J.; Leitner, W. Angew.
Chem. Int. Ed. 2012, 51, 7499-7502. (f) Whiteoak, C. J.; Kielland, N.; Laserna, V.; EscuderoAdán, E. C.; Martin, E.; Kleij, A. W. J. Am. Chem. Soc. 2013, 135, 1228-1231. (g) Yu, B.; He, L.N. ChemSusChem 2015, 8, 52-62.
(6) (a) Martín, C.; Fiorani, G.; Kleij, A. W. ACS Catal. 2015, 5, 1353-1370. (b) Comerford, J.
W.; Ingram, I. D. V.; North, M.; Wu, X. Green Chem. 2015, 17, 1966–1987. (c) Fiorani, G.; Guo,
W.; Kleij, A. W. Green Chem. 2015, 17, 1375–1389. (d) Schäffner, B.; Schäffner, F.; Verevkin,
S. P.; Börner, A. Chem. Rev. 2010, 110, 4554–4581.
(7) (a) Nozaki, K. Pure Appl. Chem. 2004, 76, 541−546. (b) Darensbourg, D. Chem. Rev. 2007,
107, 2388−2410. (c) Lu, X.-B.; Darensbourg, D. J. Chem. Soc. Rev. 2012, 41, 1462−1484. (d)
Childers, M. I.; Longo, J. M.; Van Zee, N. J.; LaPointe, A. M.; Coates, G. W. Chem. Rev. 2014,
114, 8129−8152.

23

(8) (a) Rintjema, J.; Epping, R.; Fiorani, G.; Martín, E.; Escudero-Adán, E. C.; Kleij, A. W.
Angew. Chem. Int. Ed. 2016, 55, 3972-3976. (b) Cozzolino, M.; Press, K.; Mazzeo, M.; Lamberti,
M. ChemCatChem 2015, 8, 455-460. (c) North, M.; Quek, S. C. Z.; Pridmore, N. E.; Whitwood,
A. C.; Wu, X. ACS Catal. 2015, 5, 3398-3402. (d) Tian, D.; Liu, B.; Gan, Q.; Li, H.; Darensbourg,
D. J. ACS Catal. 2012, 2, 2029-2035. (e) Rintjema, J.; Guo, W.; Martin, E.; Escudero-Adán, E. C.;
Kleij, A. W. Chem. Eur. J. 2015, 21, 10754-10762. (f) Qin, Y.; Guo, H.; Sheng, X.; Wang, X.;
Wang, F. Green Chem. 2015, 17, 2853-2858. (g) Fuchs, M. A.; Altesleben, C.; Zevaco, T. A.;
Dinjus, E. Eur. J. Inorg. Chem. 2013, 4541-4545.
(9) (a) Whiteoak, C. J.; Martin, E.; Martínez Belmonte, M.; Benet-Buchholz, J.; Kleij, A. W.
Adv. Synth. Catal. 2012, 354, 469-476. (b) Buchard, A.; Kember, M. R.; Sandeman, K. G.;
Williams, C. K. Chem. Commun. 2011, 47, 212-214. (c) Della Monica, F.; Vummaleti, S. V. C.;
Buonerba, A.; De Nisi, A.; Monari, M.; Milione, S.; Grassi, A.; Cavallo, L.; Capacchione, C. Adv.
Synth. Catal. 2016, 358, 3231-3243. (d) Buonerba, A.; De Nisi, A.; Grassi, A.; Milione, S.;
Capacchione, C.; Vagin, S.; Rieger, B. Catal. Sci. Technol. 2015, 5, 118-123.
(10) (a) A. Decortes, A. W. Kleij, ChemCatChem2011, 3, 831-834. (b) M. A. Fuchs, C.
Altesleben, S. C. Staudt, O. Walter, T. A. Zevaco, E. Dinjus, Catal. Sci. Technol. 2014, 4, 16581673. (c) J. Hu, J. Ma, Q. Zhu, Q. Qian, H. Han, Q. Mei, B. Han, Green Chem. 2016, 18, 382-385.
(d) Y. Ren, J. Chen, C. Qi, H. Jiang, ChemCatChem 2015, 7, 1535-1538. (e) Mercadé, E.;
Zangrando, E.; Claver, C.; Godard, C. ChemCatChem 2016, 8, 234-243. (f) Ren, Y.; Chen, J.; Qi,
C.; Jiang, H. ChemCatChem2015, 7, 1535-1538.
(11) For some examples: (a) Lu, X.-B.; Liang, B.; Zhang, Y.-J.; Tian, Y.-Z.; Wang, Y.-M.; Bai,
C.-X.; Wang, H.; Zhang, R. J. Am. Chem. Soc. 2004, 126, 3732–3733. (b) Paddock, R. L.; Nguyen,

24

S. T. Chem. Commun. 2004, 1622-1623. (c) Paddock, R. L.; Nguyen, S. T. J. Am. Chem. Soc.
2001, 123, 11498–11499. (d) Chang, T.; Jin, L.; Jing, H. ChemCatChem 2009, 1, 379-383. (e)
Ren, W.-M.; Wu, G.-P.; Lin, F.; Jiang, J.-Y.; Liu, C.; Luo, Y.; Lu, X.-B. Chem. Sci. 2012, 3, 2094–
2102.
(12) Decortes, A.; Castilla, A. M.; Kleij, A. W. Angew. Chem. Int. Ed. 2010, 49, 9822-9837.
(13) Licini, G.; Mba, M.; Zonta, C. Dalton Trans. 2009, 5265-5277.
(14) (a) Mba, M.; Prins, L. J.; Licini, G. Org. Lett. 2007, 9, 21-24. (b) Mba, M.; Prins, L. J.;
Zonta, C.; Cametti, M.; Valkonen, A.; Rissanen, K.; Licini, G. Dalton Trans. 2010, 39, 7384-7392.
(c) Zonta, C.; Cazzola, E.; Mba, M.; Licini, G. Adv. Synth. Catal. 2008, 350, 2503-2506.
(15) (a) Badetti, E.; Romano, F.; Marchiò, L.; Taşkesenlioğlu, S.; Daştan, A.; Zonta, C.; Licini,
G. Dalton Trans. 2016, 45, 14603-14608. (b) Romano, F.; Linden, A.; Mba, M.; Zonta, C.; Licini,
G. Adv. Synth. Catal. 2010, 352, 2937-2942.
(16) (a) Whiteoak, C. J.; Gjoka, B.; Martin, E.; Martínez Belmonte, M.; Escudero-Adán, E. C.;
Zonta, C.; Licini, G.; Kleij, A. W. Inorg. Chem. 2012, 51, 10639-10649. (b) Badetti, E.; Gjoka,
B.; Nagy, E. M.; Bernardinelli, G.; Kündig, P. E.; Zonta, C.; Licini, G. Eur. J. Inorg. Chem. 2015,
3478–3484.
(17) (a) Groysman, S.; Goldberg, I.; Goldschmidt, Z.; Kol, M. Inorg. Chem. 2005, 44, 50735080. (b) Wolff, F.; Lorber, C.; Choukroun, R.; Donnadieu, B. Inorg. Chem. 2003, 42, 7839-7845.
(18) (a) Mba, M.; Pontini, M.; Lovat, S.; Zonta, C.; Bernardinelli, G.; Kündig, P. E.; Licini, G.
Inorg. Chem. 2008, 47, 8616-8618. (b) Barroso, S.; Adão, P.; Madeira, F.; Teresa Duarte, M.;
Costa Pessoa, J.; Martins, A. M. Inorg. Chem. 2010, 49, 7452-7463. For an extensive review on

25

this topic: (c) Amadio, E.; di Lorenzo, R.; Zonta, C.; Licini, G. Coord. Chem. Rev. 2015, 301‒302,
147–162.
(19) (a) Coletti, A.; Whiteoak, C. J.; Conte, V.; Kleij, A. W. ChemCatChem 2012, 4, 1190-1196.
Recently North and coworkers used V(V) salen complexes as catalysts for the formation of cyclic
carbamates, see: (b) Beattie, C.; North, M. RSC Adv. 2014, 4, 31345-31352. Note a highly recent
contribution from Kerton and coworkers that describes the use of V(V) bisphenolate complexes
for cyclic carbonate synthesis: (c) Elkurtehi, A. I.; Kerton, F. M. ChemSusChem 2017, DOI:
10.1002/cssc.201601548.
(20) Selected examples: (a) Laserna, V.; Fiorani, G.; Whiteoak, C. J.; Martin, E.; EscuderoAdán, E. C.; Kleij, A. W. Angew. Chem. Int. Ed. 2014, 53, 10416-10419. (b) Qin, J.; Wang, P.; Li,
Q.; Zhang, Y.; Yuan, D.; Yao, Y. Chem. Commun. 2014, 50, 10952-10955. (c) Castro-Osma, J.
A.; Lamb, K. J.; North, M. ACS Catal. 2016, 6, 5012–5025. (d) Martínez-Rodríguez, L.; Otalora
Garmilla, J.; Kleij, A. W. ChemSusChem 2016, 9, 749-755. (e) Whiteoak, C. J.; Martin, E.;
Escudero-Adán, E. C.; Kleij, A. W. Adv. Synth. Catal. 2013, 355, 2233-2239. (f) Martínez, J.;
Castro-Osma, J. A.; Earlam, A.; Alonso-Moreno, C.; Otero, A.; Lara-Sánchez, A.; North, M.;
Rodríguez-Diéguez, A. Chem. Eur. J. 2015, 21, 9850–9862.
(21) Whiteoak, C. J.; Kielland, N.; Laserna, V.; Castro-Gómez, F.; Martin, E.; Escudero-Adán,
E. C.; Bo, C.; Kleij, A. W. Chem. Eur. J. 2014, 20, 2264-2275.
(22) Note that typically a bromide based nucleophile such as NBu4Br is used for the conversion
of internal epoxides. For exemplary use of bromide based nucleophiles for these types of
transformations see references 5f, 8e, 9a, 20a and 20d.

26

(23) For a highly recent and detailed benchmarking studies involving various Al(III) based
binary

catalysts

see:

Rintjema,

J.;

Kleij,

A.

W.

ChemSusChem

2017,

DOI:

10.1002/cssc.201601712.
(24) For recent examples of catalytic bio-based cyclic carbonate synthesis see: (a) Tenhumberg,
N.; Büttner, H.; Schäffner, B.; Kruse, D.; Blumenstein, M.; Werner, T. Green Chem. 2016, 18,
3775-3788. (b) Fiorani, G.; Stuck, M.; Martín, C.; Martínez-Belmonte, M.; Martin, E.; EscuderoAdán, E. C.; Kleij, A. W. ChemSusChem 2016, 9, 1304-1311. (c) Maisonneuve, L.; More, A. S.;
Foltran, S.; Alfos, C.; Robert, F.; Landais, Y.; Tassaing, T.; Grau, E.; Cramail, H. RSC Adv. 2014,
4, 25795-25803. (d) Langanke, J.; Greiner, L.; Leitner, W. Green Chem. 2013, 15, 1173-1182.
(25) Recently a H2O-ligated hexa-coordinated V(V) aminotriphenolate complex was
crystallographically and spectroscopically characterized showing that a coordination site is indeed
available for O‒donor ligands, see reference 17a.
(26) Since the 2‒PO complex is sterically more congested than complex 2 itself, we decided to
use NBu4Br rather than NBu4I for further catalytic tests as bromide is a smaller sized nucleophile.
Note that for Al(III) based binary catalysts we found that the use bromide and iodide based
nucleophiles induced similar type of activity in terminal epoxide conversions, see reference 5f and
21.
(27) The Xray molecular structure for complex 1 was reported before, see reference 17a.
(28) In order to support the fast formation of 2‒PO from 2 and PO, we carried out a timedependent

51

V NMR experiment (50ºC) combining the V-complex 2 and a large excess of 200

molar equiv of PO in CD3CN. Despite the fact that the dilution is much higher than reported in
Table 2 (reactions carried out “neat”), smaller excess of PO was used (200 versus 833 equiv) and

27

the lack of efficient stirring during the NMR experiment, we found that the formation of 2‒PO is
relatively fast with around 50% of the V-complex being present as 2‒PO after 1.5 h. For details
see the SI.

28

Insert Table of Contents Graphic and Synopsis Here

29